초록

A system and method for determining which of a plurality of pre-data-collection software applications to apply to optically imaged responses received from a response provider.

대표청구항 ▼

A system and method for determining which of a plurality of pre-data-collection software applications to apply to optically imaged responses received from a response provider. cation of Artificial Machine Intelligence", Am J Clin Pathol, 1991, pp. 134-141, vol. 96. D. V. Cicchetti, "Neural Networks

A system and method for determining which of a plurality of pre-data-collection software applications to apply to optically imaged responses received from a response provider. cation of Artificial Machine Intelligence", Am J Clin Pathol, 1991, pp. 134-141, vol. 96. D. V. Cicchetti, "Neural Networks and Diagnosis in the Clinical Laboratory: State of the Art", Clin. Chem., 1992, pp. 9-10, vol. 38, No. 1. R. Ashfaq, et al., "Evaluation of PAPNET� System for Rescreening of Negative Cervical Smears", Diagnostic Cytopathology, 1995, pp. 31-36, vol. 13, No. 1. D. C. Malins, et al., "Models of DNA structure achieve almost perfect discrimination between normal prostate, benign prostatic hyperplasia (BPH), and adenocarcinoma and have a high potential for predicting BPH and prostase cancer", Proc. Natl. Acad. Sci. USA, Jan. 1997, pp. 259-264, vol. 94. I. W. Ricketts, et al., "Towards the Automated Prescreening of Cervical Smears", IEE Colloquium on Applications of Image Processing in Mass Health Screening, Digest No. 056, Mar. 11, 1992, pp. 7/1-7/4. H. Kohno, et al., "Quantitative Analysis of Scintiscan Matrices by Computer", Japanese Journal of Medical Electronics and Biological Engineering, Aug. 1974, pp. 218-225, vol. 12, No. 4. Salford Systems White Paper Series, http://www.salford-systems.com/whitepaper.html, printed Oct. 17, 2000. V. Berikov, et al., "Regression trees for analysis of mutational spectra in nucleotide sequences", Bioinformatics, 1999, pp. 553-562, vol. 15, Nos. 7/8. L. Breiman, et al., Chapters 6-8 in Classification and Regression Trees, CRC Press (Boca Raton), 1998, pp. 174-265. J. M. Halket, et al., "Deconvolution Gas Chromatography/Mass Spectrometry of Urinary Organic Acids--Potential for Pattern Recognition and Automated Identification of Metabolic Disorders", Rapid Commun. Mass Spectrom, 1999, pp. 279-284, vol. 13. A. Eghbaldar, et al., "Identification of Structural Features from Mass Spectrometry Using a Neural Network Approach: Application of Trimethylsilyl Derivatives Used for Medical Diagnosis", J. Chem. Inf. Comput. Sci., 1996, pp. 637-643, vol. 36. R. J. Babaian, et al., "Performance of a Neural Network in Detecting Prostate Cancer in the Prostate-Specific Antigen Reflex Range of 2.5 to 4.0 ng/mL", Urology, 2000, pp. 1000-1006, vol. 56, No. 6. C. S. Tong, et al., "Mass Spectral Search method using the Neural Network approach", Proceedings, International Joint Conference on Neural Networks, Washington, DC, Jul. 1999. pp. 3962-3967, vol. 6. C. S. Tong, et al., "Mass spectral search method using the neural network approach", Chemometrics and Intelligent Laboratory Systems, 1999, pp. 135-150, vol. 49. R. R. Hashemi, et al., "Identifying and Testing of Signatures for Non-Volatile Biomolecules Using Tandem Mass Spectra", Sigbio newsletter, ACM Press, Dec. 1995, pp. 11-19, vol. 15, No. 3. I. Belic, et al., "Neural network methodologies for mass spectra recognition", Vacuum, 1997, pp. 633-637, vol. 48, Nos. 7-9. W. Werther, et al., "Classification of mass spectra, A comparison of yes/no classification methods for the recognition of simple structural properties", Chemometrics and Intelligent Laboratory Systems, 1994, pp. 63-76, vol. 22. A. Y. Cairns, et al., "Towards the Automated Prescreening of Breast X-Rays", Digest of the IEE Colloquium, Applications of Image Processing in Mass Health Screening, University of Dundee, pp. 1/1-1/5. M. Astion, et al., The Application of Backpropagation Neural Networks to Problems in Pathology and Laboratory Medicine, Arch Pathol Lab Med, Oct. 1992, pp. 995-1001, vol. 116. R. Goodacre, "Rapid identification of urinary tract infection bacteria using hyperspectral whole-organism fingerprinting and artificial neutral networks", Microbiology, 1998, pp. 1157-1170, vol. 144. J. Taylor, "The deconvolution of pyrolysis mass spectra using genetic programming: application to the identification of some Eubacterium species", FEMS Microbiology Letters, 1998, pp. 237-246, vol. 160. R. Goodacre, et al., Discrimination between methicillin-resistant and methicillin-susceptible Staphylococcus aureus using pyrolysis mass spectrometry and artificial ne utral networks, Journal of Antimicrobial Chemotherapy, 1998, pp. 27-34, vol. 41. J. Chun, et al., "Long-term Identification of Streptomycetes Using Pyrolysis Mass Spectrometry and Artificial Neural Networks", Zbl. Bakt., 1997, pp. 258-266, vol. 285. R. G. W. Kenyon, et al., "Application of Neural Networks to the Analysis of Pyrolysis Mass Spectra", Zbl. Bakt., 1997, pp. 267-277, vol. 285. T. Nilsson, et al., "Classification of Species in the Genus Penicillium by Curie Point Pyrolysis/Mass Spectrometry Followed by Multivariate Analysis and Artificial Neural Networks", Journal of Mass Spectrometry, 1996, pp. 1422-1428, vol. 31. R. Goodacre, et al., "Sub-species Discrimination, Using Pyrolysis Mass Spectrometry and Self-organising Neural Networks, of Propionibacterium acnes Isolated from Normal Human Skin", Zbl. Bakt., 1996, pp. 501-515, vol. 284. R. Goodacre, et al., "Quantitative Analysis of Multivariate Data Using Artificial Neural Networks: A Tutorial Review and Applications to the Deconvolution of Pyrolysis Mass Spectra", Zbl. Bakt., 1996, pp. 516-539, vol. 284. R. Goodacre, et al., "Identification and Discrimination of Oral Asaccharolytic Eubacterium spp. by Pyrolysis Mass Spectrometry and Artificial Neural Networks", Current Microbiology, 1996, pp. 77-84, vol. 32. R. Goodacre, et al., "Correction of Mass Spectral Drift Using Artificial Neural Networks", Anal. Chem., 1996, pp. 271-280, vol. 68. R. Freeman, et al., "Resolution of batch variations in pyrolysis mass spectrometry of bacteria by the use of artificial neural network analysis", Antonie van Leeuwenhoek, 1995, pp. 253-260, vol. 68. D. H. Chace, et al., "Laboratory integration and utilization of tandem mass spectrometry in neonatal screening: a model for clinical mass spectrometry in the next millennium", Acta Paediatr Supp 432, 1999, pp. 45-47, vol. 88. B. Curry, et al., "MSnet: A Neural Network That Classifies Mass Spectra", Stanford Science Center, Stanford University, Stanford, California, Oct. 1990, pp. 1-31. R. A. Shaw, et al., "Infrared Spectroscopy of Exfoliated Cervical Cell Specimens", Analytical and Quantitative Cytology and Histology, Aug. 1999, pp. 292-302, vol. 21, No. 4. I. Belic, "Neural Networks Methodologies for Mass Spectra Recognition", 4 pgs. C. Prior, et al., "Potential of Urinary Neopterin Excretion in Differentiating Chronic Non-A, Non-B Hepatitis From Fatty Liver", The Lancet, Nov. 1987, pp. 1235-1237. John R. Yates, III, et. al., "Mass Spectrometry and the Age of the Proteome", Journal of Mass Spectrometry, 1998, pp. 1-19, vol. 33. Arno Hausen, et al., "Determination of Neopterine in Human Urine by Reversed-Phase High-Performance Liquid Chromatography", Journal of Chromatography, 1982, pp. 61-70, vol. 227. Andrej Shevchenko, et al., "MALDI Quadrupole Time-of-Flight Mass Spectrometry: A Powerful Tool for Proteomic Research", Anal. Chem., 2000, pp. 2132-2141, vol. 72, No. 9. Cloud P. Paweletz, et al., "Rapid Protein Display Profiling of Cancer Progression Directly From Human tissue Using a Protein Biochip", Drug Development Research, 2000pp. 34-42, vol. 49. Anil K. Jain, et al., "Statistical Pattern Recognition: A Review", IEEE Transactions on Pattern Analysis and Machine intelligence, Jan. 2000, pp. 4-37, vol. 22, No. 1. Sandrine Dudoit, et al., "Comparison of Discrimination Methods for the Classification of Tumors Using Gene Expression Data", Technical report #576, Jun. 2000, pp. 1-43. Dudoit, S., et al., "Comparison of discrimination methods for the classification of tumors using gene expression data", UC Berkeley, Slides, 52 pages, URL=http://stat-www.berkeley.edu/users/terry/zarray/Html/discr.html, (Mar. 7, 2000). Nikulin, Alexander E., et al., "Near-optimal region selection for feature space reduction: novel preprocessing methods for classifying MR spectra", NMR in Biomedicine, (1998), 209-216, vol. 11. Alaiya et al., "Classification of Human Ovarian Tumors Using Multivariate Data Analysis of Polypeptide Expression Patterns." Int. J. Cancer, vol. 86, pp. 731-736, Wiley-Liss, Inc., (2000). Bailey-Kellogg et al., "Reducing Mass Degeneracy in SAR by MSby Stable Isotopic Labeling." Journal of Computational Biology, vol. 8, No. 1, pp 19-36, Mary Ann Liebert, Inc., (2001). Caprioli et al. "Molecular Imaging of Biological Samples: Localization of Peptides and Proteins Using MALDI-TOF MS." Analytical Chemistry, vol. 69, No. 23, pp 4751-4760, American Chemical Society, (Dec. 1, 1997). George, "A Visualization and Design Tool (AVID for Data Mining with the Self-Organizing Feature Map." International Journal on Artificial Intelligence Tools, vol. 9, No. 3, pp. 369-375, World Scientific Publishing Company, (2000). Kohavi et al., "Wrappers for Feature Subset Selection." Artificial Intelligence, vol. 97, No. 1-2, pp. 273-324, Elsevier Science B.V., (1997). Marvin et al., "Characterization of a novel Sepia officinalis neuropeptide using MALDI-TOF MS and post-source decay analysis." Peptides, vol. 22, No. 9., pp 1391-1396, Elsevier Science Inc., (Sep. 2001). Oh et al., "A Database of Protein Expression in Lung Cancer." Proteomics, 1, pp. 1303-1319, WILEY-VCH Verlag GmbH, (2001). Strouthopoulos et al., "PLA using RLSA and a Neural Network." Engineering Applications of Artificial Intelligence, vol. 12, No. 2, pp. 119-138, Elsevier Science Ltd., (1999). Taylor et al., "The Deconvolution of Pyrolysis Mass Spectra using Genetic Programming: Application to the Identification of Some Eubacterium Species." FEMS Microbiology Letters, 160, pp. 237-246, Elsevier Science B.V., (1998). Zhang, "Combining Multiple Biomarkers in Clinical Diagnostics--A Review of Methods and Issues." pp. 1-14. sification model comprises using pre-existing marker data to form the classification model. 14. The method of claim 1 wherein the data set is formed by: detecting signals in the mass spectra, each mass spectrum comprising data representing signal strength as a function of mass-to-charge ratio; clustering the signals having similar mass-to-charge ratios into signal clusters; selecting signal clusters having at least a predetermined number of signals with signal intensities above a predetermined value; identifying the mass-to-charge ratios corresponding to the selected signal clusters; and forming the data set using signal intensities at the identified mass-to-charge ratios. 15. The method of claim 1 wherein forming the classification model comprises at least one of identifying features that discriminate between the different biological statuses, and learning. 16. The method of claim 1 wherein the classification process is a binary recursive partitioning process. 17. The method of claim 1 further comprising: c) interrogating the classification model to determine if one or more features discriminate between the different biological statuses. 18. The method of claim 1 further comprising: c) repeating a) and b) using a larger plurality of samples. 19. The method of claim 1 wherein the mass spectra are derived from a surface enhanced laser desorption/ionization process using a substrate comprising an affinity material, wherein the affinity material comprises antibodies. 20. A method for classifying an unknown sample into a class characterized by a biological status using a digital computer, the method comprising: a) entering data obtained from a mass spectrum of the unknown sample into a digital computer, wherein the mass spectrum is derived from a surface enhanced laser desorption/ionization process using a substrate comprising an affinity material, wherein the affinity material comprises antibodies; and b) processing the mass spectrum data using the classification model formed by the method of claim 1 to classify the unknown sample in a class characterized by a biological status. 21. The method of claim any of claims 1, 2, and 6-11 wherein each mass spectrum comprises data representing signal strength as a function of mass-to-charge ratio. 22. The method of any of claims 2, and 6-11 wherein the data set is formed by: detecting signals in the mass spectra, each mass spectrum comprising data representing signal strength as a function of mass-to-charge ratio; clustering the signals having similar mass-to-charge ratios into signal clusters; selecting signal clusters having at least a predetermined number of signals with signal intensities above a predetermined value; identifying the mass-to-charge ratios corresponding to the selected signal clusters; and forming the data set using signal intensities at the identified mass-to-charge ratios. 23. A method that analyzes mass spectra using a digital computer, the method comprising: a) entering into the digital computer a data set obtained from mass spectra from a plurality of samples, wherein each sample is, or is to be assigned to a class within a class set comprising two or more classes, each class characterized by a different biological status, and wherein each mass spectrum comprises data representing signal strength as a function of time-of-flight, mass-to-charge ratio, or a value derived from time-of-flight or mass-to-charge ratio; and b) forming a classification model which discriminates between the classes in the class set, wherein forming comprises analyzing the data set by executing code that embodies a classification process comprising a recursive partitioning process, and wherein the method further comprises forming the data set, wherein forming the data set comprises obtaining raw data from the mass spectra and then preprocessing the raw mass spectra data to form the data set. 24. The method of claim 1 wherein the different classes are selected from exposure to a drug, exposure to one of a class of drugs and lack of exposure to a drug or one of a class of drugs. 25. The method of claim 1 wherein the each mass spectrum comprises data representing signal strength as a function mass-to-charge ratio or a value derived from mass-to-charge ratio. 26. A method for classifying an unknown sample into a class characterized by a biological status using a digital computer, the method comprising: a) entering data obtained from a mass spectrum of the unknown sample into a digital computer; and b) processing the mass spectrum data using the classification model formed by the method of claim 1 to classify the unknown sample in a class characterized by a biological status. 27. The method of claim 26 wherein the different biological statuses comprise un-diseased, low grade cancer and high grade cancer. 28. The method of claim 26 wherein the class is characterized by exposure to a drug of one of a class of drugs. 29. The method of claim 26 wherein the class is characterized by response to a drug. 30. The method of claim 26 wherein the class is characterized by a toxicity status. 31. A method for estimating the likelihood that an unknown sample is accurately classified as belonging to a class characterized by a biological status using a digital computer, the method comprising: a) entering data obtained from a mass spectrum of the unknown sample into a digital computer; and b) processing the mass spectrum data using the classification model formed by the method of claim 1 to estimate the likelihood that the unknown sample is accurately classified into a class characterized by a biological status. 32. A computer readable medium comprising: a) code for entering data obtained from a mass spectrum of an unknown sample into a digital computer; and b) code for processing the mass spectrum data using the classification model formed by the method of claim 1 to classify the unknown sample in a class characterized by a biological status. 33. A system comprising: a gas phase ion spectrometer; a digital computer adapted to process data from the gas phase ion spectrometer; and the computer readable medium of claim 32 in operative association wit the digital computer. 34. The system of claim 33 wherein the gas phase ion spectrometer is adapted to perform a laser desorption ionization process. 35. A computer readable medium comprising: a) code for entering data obtained from a mass spectrum of an unknown sample into a digital computer; and b) code for processing the mass spectrum data using the classification model formed by the method of claim 1 to estimate the likelihood that the unknown sample is accurately classified into a class characterized by a biological status. 36. A system comprising: a gas phase ion spectrometer; a digital computer adapted to process data from the gas phase ion spectrometer; and the computer readable medium of claim 35 in operative association with the digital computer. 37. The system of claim 36 wherein the gas phase ion spectrometer is adapted to perform a laser desorption ionization process. 38. The method of claim 23 wherein the mass spectra are selected from the group consisting of MALDI spectra, surface enhanced laser desorption/ionization spectra, and electrospray ionization spectra. 39. The method of claim 23 wherein the class set consists of exactly two classes. 40. The method of claim 23 wherein the samples comprise biomolecules selected from the group consisting of polypeptides and nucleic acids. 41. The method of claim 23 wherein the samples are derived from a eukaryote, a prokaryote or a virus. 42. The method of claim 23 wherein the different biological statuses comprise a normal status and a pathological status. 43. The method of claim 23 where the different biological statuses comprise un-diseased, low grade cancer and high grade cancer. 44. The method of claim 23 wherein the different biological statuses compris